use "SovietCollaboration_MainDataset.dta"
xi: gen i.year i.adjustedmsc
***New Mahdi- Try 1980 as the cut-off point
set more off

drop if sovietauthor==1
drop if newbdrankingposition>3 & newbdrankingposition<31
gen topsovietdummy=0
replace topsovietdummy=1 if newbdrankingposition<4
gen after1980 =0
replace after1980=1 if year>1980
gen topsovietafter1980=topsovietdummy*after1980
xtset adjustedmsc
xtreg  logauthourcount  topsovietafter1980  _I*, robust cluster(adjustedmsc)
outreg using alternate.doc, merge
*Drop after 1990 so we have a balanced time span
drop if year>1990
xtset adjustedmsc
xtreg  logauthourcount  topsovietafter1980 _I*, robust cluster(adjustedmsc)
outreg using alternate.doc, merge

*****New Mahdi- try 2000 as cut-off
clear
use "SovietCollaboration_MainDataset.dta"
set more off
drop if sovietauthor==1
drop if newbdrankingposition>3 & newbdrankingposition<31
gen topsovietdummy=0
replace topsovietdummy=1 if newbdrankingposition<4
gen after2000 =0
replace after2000=1 if year>2000
gen topsovietafter2000=topsovietdummy*after2000
xtset adjustedmsc
xtreg  logauthourcount  topsovietafter2000  _I*, robust cluster(adjustedmsc)
outreg using alternate.doc, merge

**Check to see what happens if we drop after
drop if year<1990
xtset adjustedmsc
xtreg  logauthourcount  topsovietafter2000  _I*, robust cluster(adjustedmsc)
outreg using alternate.doc, merge

****
